NELSON GOODMAN
285
THREE TYPES OF REALISM
A typical English eighteenth-century portrait is more real–
istic, according to prevalent Western usage, than a Picasso showing
three sides of a head at once or an EI Greco accentuating upward
swirls; and a sketch of the Piazza della Signoria drawn according to
standard rules of perspective is more realistic than another drawn
according to reversed or otherwise transformed rules . Greater
realism here does not imply greater accuracy or informativeness .
The two sketches may convey exactly the same information , and the
Picasso and El Greco supply information the English portrait does
not. Moreover, a sketch drawn generally in standard perspective but
with several errors is more realistic than a sketch in reversed perspec–
tive with no errors; and the pictures that sometimes appear as games
with such captions as "Find fifteen mistakes in this picture" are in–
variably realistic. Realism in all these cases depends upon familiar–
ity; the pictures in the accustomed, standard mode of representation
count as the more realistic.
"Realism", though, has another use as well . Practice palls; and
a new mode of representation may be so fresh and forceful as to achieve
what amounts to a revelation
(Languages ojArt,
pp.
37-38).
This was
true for standard Western perspective when it was invented during
the Renaissance, and no less true for modes that broke away from
that system, such as the Oriental mode when rediscovered by late–
nineteenth-century French painters, and various modes developed
by later artists.
Although realism thus may sometimes be associated with
revelation, this should not be taken to imply that representation of
any sort consists of faithful reporting on 'the real world'. For there is ,
I maintain, no such thing as the real world, no unique, ready-made ,
absolute reality apart from and independent of all versions and vi–
sions. Rather, there are many right world,versions, some of them ir–
reconcilable with others; and thus there are many worlds if any. A
version is not so much made right by a world as a world is made by a
right version. Obviously rightness has therefore to be determined
otherwise than by matching a version with a world. My relativism,
which nevertheless recognizes the difference between right and
wrong versions, does not stop with representations and vision and
realism and resemblance but goes through to reality as well.
